Stimulant induced psychosis is one of the major health sequelae of stimulant abuse. It resembles manic psychosis, and likely shares some aspects of a common dopamine pathophysiology with endogenous psychoses. The imminent completion of the project to identify all human and rodent genes provides a powerful tool for the comprehensive study of pathophysiological processes at the genetic level. Microarray technology will soon enable the simultaneous study of expression of all genes. We propose to employ this new technology in order to identify novel candidate genes that play a critical role in the pathophysiology of amphetamine psychosis and endogenous psychoses. We have recently described a rat model of amphetamine psychosis which involves the administration of escalating doses followed by binges in order to mimic patterns of human abuse that result in psychosis. This escalating dose binge (EDB) paradigm results in a distinct behavioral response which we hypothesize is analogous to human psychosis. The challenge with this new technology is the interpretation of the large amounts of data it generates. Genes of interest will be identified by the convergence of several types of criteria: time course, administration paradigm and chromosomal localization. Rats will be administered amphetamine or saline according to the EDB paradigm and sacrificed 24 hours after the last dose. Five brain regions implicated in amphetamine psychosis and one control region will be examined. The RNA levels for 24,000 genes will be determined in individual animals using Affymetrix GeneChip technology. As amphetamine psychosis is a chronic phenomenon, the 24 hour delay will reduce the number of changed genes to those of greater detail using a time course and in situ hybridization. We hypothesize that genes relevant to the mechanism of amphetamine psychosis will show increasing change through the course of the EDB paradigm, and that their map position in the human genome will correspond to known linkage peaks for bipolar disorder or schizophrenia. Our preliminary studies already suggest the power of this convergent approach. In animals treated with a single dose of amphetamine the gene with highest level of expression in prefrontal cortex was identical to a positional candidate on 22q identified in human linkage studies of bipolar disorder.